1 1 Introduction This book is not about corrosion; rather it is about paints that prevent corrosion. It was written for those who must protect structural steel from rusting by using anti- corrosion paints. The philosophy of this book is this: if one knows enough about paint, one need not be an expert on rust. In keeping with that spirit, the book endeavors to cover the field of heavy-duty anticorrosion coatings without a single anode or cathode equation explaining the corrosion process. It is enough for us to know that steel will rust if allowed to; we will concentrate on preventing it. 1.1 SCOPE OF THE BOOK The scope of this book is heavy-duty protective coatings used to protect structural steel, infrastructure components made of steel, and heavy steel process equipment. The areas covered by this book have been chosen to reflect the daily concerns and choices faced by maintenance engineers who use heavy-duty coating, including: • Composition of anticorrosion coatings • Waterborne coatings • Blast-cleaning and other heavy surface pretreatments • Abrasive blasting and heavy-metal contamination • Weathering and aging of paint • Corrosion testing — background and theoretical considerations • Corrosion testing — practice 1.1.1 T ARGET G ROUP D ESCRIPTION The target group for this book consists of those who specify, formulate, test, or do research in heavy-duty coatings for such applications as: • Boxes and girders used under bridges or metal gratings used in the decks of bridges • Poles for traffic lights and street lighting • Tanks for chemical storage, potable water, or waste treatment • Handrails for concrete steps in the fronts of buildings • Masts for telecommunications antennas • Power line pylons • Beams in the roof and walls of food-processing plants • Grating and framework around processing equipment in paper mills 7278_C001.fm Page 1 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC 2 Corrosion Control Through Organic Coatings All of these forms of structural steel have at least two things in common: 1. Given a chance, the iron in them will turn to iron oxide. 2. When the steel begins rusting, it cannot be pulled out of service and sent back to a factory for treatment. During the service life of one of these structures, maintenance painting will have to be done on-site. This imposes certain limitations on the choices the maintenance engineer can make. Coatings that must be applied in a factory cannot be reapplied once the steel is in service. This eliminates organic paints, such as powder coatings or electrodeposition coatings, and several inorganic pretreatments, such as phosphat- ing, hot-dip galvanizing, and chromating. New construction can commonly be pro- tected with these coatings, but they are almost always a one-time-only treatment. When the steel has been in service for a number of years and maintenance coating is being considered, the number of practical techniques is narrowed. This is not to say that the maintenance engineer must face corrosion empty-handed; more good paints are avail- able now than ever before, and the number of feasible pretreatments for cleaning steel in-situ is growing. In addition, coatings users now face such pressures as environmental responsibility in choosing new coatings and disposing of spent abrasives as well as increased awareness of health hazards associated with certain pretreatment methods. 1.1.2 S PECIALTIES O UTSIDE THE S COPE Certain anticorrosion coating subspecialties fall outside the scope of this work, includ- ing those dealing with automotive, airplane, and marine coatings; powder coatings; and coatings for cathodic protection. These methods are all economically important and scientifically interesting but lie outside of our target group for one or more reasons: • The way in which the paint is applied can be done only in a factory, so maintenance painting in the field is not possible. (Automotive and powder coatings) • Aluminium — not steel —is used as the substrate, and the coatings experience temperature extremes and ultraviolent loads that earth-bound structures and their coatings never encounter. (Airplane coatings) • The circumstances under which marine coatings and coatings with cathodic protection must operate are so different from those experienced by the infra- structure in the target group that different coating and testing technologies are needed. These exist and are already well covered in the technical literature. 1.2 PROTECTION MECHANISMS OF ORGANIC COATINGS This section presents a brief overview of the various mechanisms by which organic coatings provide corrosion protection to the metal substrate. Corrosion of a painted metal requires all of the following elements [1]: • Water • Oxygen or another reducible species 7278_C001.fm Page 2 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC Introduction 3 • A dissolution process at the anode • A cathode site • An electrolytic path between the anode and cathode Any of these items could potentially be rate controlling. A coating that can suppress one or more of the items listed above can therefore limit the amount of corrosion. The main protection mechanisms used by organic coatings are: • Creating an effective barrier against the corrosion reactants water and oxygen • Creating a path of extremely high electrical resistance, thus inhibiting anode-cathode reactions • Passivating the metal surface with soluble pigments • Providing an alternative anode for the dissolution process The last two protection mechanisms listed above are discussed extensively in Chapter 2. This section will therefore concentrate on the first two protection mechanisms in the list above. It must be noted that it is impossible to use all these mechanisms in one coating. For example, pigments whose dissolved ions passivate the metal surface require the presence of water. This rules out their use in a true barrier coating, where water penetration is kept as low as possible. In addition, the usefulness of each mechanism depends on the service environ- ment. Guruviah studied corrosion of coated panels under various accelerated test methods with and without sodium chloride (salt). Where salt was present, electrolytic resistance of the coatings was the dominant factor in predicting performance. How- ever, in a generally similar method with no sodium chloride, oxygen permeation was the rate-controlling factor for the same coatings [2]. 1.2.1 D IFFUSION OF W ATER AND O XYGEN Most coatings, except specialized barrier coatings such as chlorinated rubber, do not protect metal substrates by preventing the diffusion of water. The attractive force for water within most coatings is simply too strong. There seems to be general agreement that the amount of water that can diffuse through organic coatings of reasonable thickness is greater than that needed for the corrosion process [2–8]. Table 1.1 shows the permeation rates of water vapor through several coatings as measured by Thomas [9,10]. The amount of water necessary for corrosion to occur at a rate of 0.07 g Fe/cm 2 /year is estimated to be 0.93 g/m 2 /day [9,10]. Thus, coatings with the lowest permeability rates might possibly be applied in sufficient thickness such that water does not reach the metal in the amounts needed for corrosion. Other coatings must provide protection through other mechanisms. Similar results have been obtained by other studies [2,11]. However, the role of water permeation through the coating cannot be completely ignored. Haagan and Funke have pointed out that, although water permeability is not normally the rate-controlling step in corrosion, it may be the rate-determining factor in adhesion loss [11]. 7278_C001.fm Page 3 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC 4 Corrosion Control Through Organic Coatings The amount of oxygen required for a corrosion rate of 0.07 g Fe/cm 2 /year is estimated to be 575 cc/m 2 /day. Thomas studied oxygen permeation rates for several types of coatings and found that they have rates far below what is needed to maintain the corrosion reaction, as shown in Table 1.2 [9,10]. These measurements were taken using 1 atmosphere of pure oxygen — that is, nearly five times the amount of oxygen available in air. In Earth’s atmosphere, oxygen transport rates may be expected to be lower than this [12]. It should perhaps be noted that these were measurements of oxygen gas permeating through the coating. The amount of oxygen reaching the metal surface will be higher, because water carries dissolved oxygen with it when permeating the coating. In general, water and oxygen are necessary for the corrosion process; however, their permeation through the coating is not a rate-determining step [13–15]. TABLE 1.1 Water Vapor Permeability Coating Type Water Vapor Permeability, g/m 2 /25µm/day Chlorinated rubber 20 ± 3 Coat tar epoxy 30 ± 1 Aluminium epoxy mastic 42 ± 6 Red-lead oil-based 214 ± 3 White alkyd 258 ± 6 Sources: Thomas. N.L., Prog. Org. Coatings , 19, 101, 1991; Thomas, N.L., Proc. Symp. Advances in Corrosion Protection by Organic Coatings, Elec- trochem. Soc. , 1989, 451. TABLE 1.2 Oxygen Permeability Coating Type Oxygen Permeability, cc/m 2 /100µm/day Chlorinated rubber 30 ± 7 Coat tar epoxy 213 ± 38 Aluminium epoxy mastic 110 ± 37 Red-lead oil-based 734 ± 42 White alkyd 595 ± 49 Sources : Thomas. N.L., Prog. Org. Coatings , 19, 101, 1991; Thomas, N.L., Proc. Symp. Advances in Corrosion Protection by Organic Coatings, Electrochem. Soc. , 1989, 451. 7278_C001.fm Page 4 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC Introduction 5 1.2.2 E LECTROLYTIC R ESISTANCE Perhaps the single most important corrosion-protection mechanism of organic coatings is to create a path of extremely high electrical resistance between anodes and cathodes. This electrical resistance reduces the flow of current available for anode-cathode corrosion reactions. In other words, water — but not ions — may readily permeate most coatings. Therefore, the water that reaches a metal substrate is relatively ion- free [12]. Steel corrodes very slowly in pure water, because the ferrous ions and hydroxyl ions form ferrous hydroxide (Fe(OH) 2 ). Fe(OH) 2 has low solubility in water (0.0067 g/L at 20 ° C), precipitates at the site of corrosion, and then inhibits the diffusion necessary to continue corrosion. On the other hand, if chloride or sulphate ions are present, they react with steel to form ferrous chloride and sulphate complexes. These are soluble and can diffuse away from the site of corrosion. After diffusing away, they can be oxidized, hydrolyzed, and precipitated as rust some distance away from the corrosion site. The stimulating Cl – or SO 4 2– anion is liberated and can re-enter the corrosion cycle until it becomes physically locked up in insoluble corrosion products [16-21]. This mechanism of blocking ions has several names, including electrolytic resistance, resistance inhibition, and ionic resistance. The terms electrolytic resistance and ionic resistance are used more-or-less interchange- ably, because Kittleberger and Elm showed a linear relationship between the diffu- sion of ions and the reciprocal of the film resistance [22]. Overall, the electrolytic resistance of an immersed coating can be said to depend on at least two factors: the activity of the water in which the coating is immersed and the nature of the counter ion inside the polymer [1]. Bacon and colleagues have performed extensive work establishing the correlation between electrolytic (ionic) resistance of the coating and its ability to protect the steel substrate from corrosion. In a study involving more than 300 coating systems, they observed good corrosion protection in coatings that could maintain a resistance of 108 Ω /cm 2 over an exposure period of several months; they did not observe the same results in coatings whose resistance fell below this [23]. Mayne deduced the importance of electrolytic resistance as a protection mech- anism from the high rates of water and oxygen transport through coatings. Specif- ically, Mayne and coworkers [7, 24-27] found that the resistance of immersed coatings could change over time. From their studies, they concluded that at least two processes control the ionic resistance of immersed coatings: • A fast change, which takes place within minutes of immersion • A slow change, which takes weeks or months [26] The fast change is related to the amount of water in the film. Its controlling factor is osmotic pressure. The slow change is controlled by the concentration of electrolytes in the immersion solution. An exchange of cations in the electrolyte for hydrogen ions in the coating may lie behind this steady fall, over months, in the coating resistance. This theory has received some support from the work of Khullar and Ulfvarson, who found an inverse relationship between the ion exchange capacity and the corrosion protection efficiency of paint films [13, 28]. The structural changes brought about by this ion exchange might slowly destroy the protective properties of the film [29]. 7278_C001.fm Page 5 Wednesday, March 1, 2006 10:54 AM © 2006 by Taylor & Francis Group, LLC 6 Corrosion Control Through Organic Coatings Many workers in the field of water transport have concentrated on the physical properties of film, such as capillary structure, or composition of the electrolytes. The work of Kumins and London has shown that the chemical composition of the polymer is equally important. In particular, the concentration of fixed anions in the polymer film is critical. They found that if the concentration of salt in the electrolyte was below the film’s fixed-anion concentration, the passage of anions through the film was very restricted. If the electrolyte’s concentration was above the polymer’s fixed- anion concentration, anions could permeate much more freely through the film [30]. Further information regarding the mechanisms of ion transport through the coating film can be found in reviews by Koehler, Walter, and Greenfield and Scantlebury [1, 29, 31]. 1.2.3 A DHESION When a metal substrate has corroded, the paint no longer adheres to it. Accordingly, corrosion workers commonly place heavy emphasis on the importance of adhesion of the organic coating to the metal substrate, and a great deal of energy has gone into developing test methods for quantifying this adhesion. 1.2.3.1 What Adhesion Accomplishes Very strong adhesion can help suppress corrosion by resisting the development of corrosion products, hydrogen evolution, or water build-up under the coating [32-35]. In addition, by bonding to as many available active sites on the metal surface as possible, the coating acts as an electrical insulator, thereby suppressing the formation of anode-cathode microcells among inhomogeneities in the surface of the metal. The role of adhesion is to create the necessary conditions so that corrosion- protection mechanisms can work. A coating cannot passivate the metal surface, create a path of extremely high electrical resistance at the metal surface, or prevent water or oxygen from reaching the metal surface unless it is in intimate contact — at the atomic level — with the surface. The more chemical bonds between the surface and coating, the closer the contact and the stronger the adhesion. An irreverent view could be that the higher the number of sites on the metal that are taken up in bonding with the coating, the lower the number of sites remaining available for electrochemical mischief. Or as Koehler expressed it: The position taken here is that from a corrosion standpoint, the degree of adhesion is in itself not important. It is only important that some degree of adhesion to the metal substrate be maintained. Naturally, if some external agency causes detachment of the organic coating and there is a concurrent break in the organic coating, the coating will no longer serve its function over the affected area. Typically, however, the detachment occurring is the result of the corrosion processes and is not quantitatively related to adhesion [1]. In summary, good adhesion of the coating to the substrate could be described as a “necessary but not sufficient” condition for good corrosion protection. For all of the protection mechanisms described in the previous sections, good adhesion of the coating 7278_C001.fm Page 6 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC Introduction 7 to the metal is a necessary condition. However, good adhesion alone is not enough; adhesion tests in isolation cannot predict the ability of a coating to control corrosion [36]. 1.2.3.2 Wet Adhesion A coating can be saturated with water, but if it adheres tightly to the metal, it can still prevent sufficient amounts of electrolytes from collecting at the metal surface for the initiation of corrosion. How well the coating clings to the substrate when it is saturated is known as wet adhesion . Adhesion under dry conditions is probably overrated; wet adhesion, on the other hand, is crucial to corrosion protection. Commonly, coatings with good dry adhesion have poor wet adhesion [37-41]. The same polar groups on the binder molecules that create good dry adhesion can wreak mischief by decreasing water resistance at the coating-metal interface — that is, they decrease wet adhesion [42]. Another important difference is that, once lost, dry adhesion cannot be recovered. Loss of adhesion in wet conditions, on the other hand, can be reversible, although the original dry adhesion strength will probably not be obtained [16, 43]. Perhaps it should be noted that wet adhesion is a coating property and not a failure mechanism. Permanent adhesion loss due to humid or wet circumstances also exists and is called water disbondment. Relatively little research has been done on wet adhesion phenomena. Leidheiser has identified some important questions in this area [43]: 1. How can wet adhesion be quantitatively measured while the coating is wet? 2. What is the governing principle by which water collects at the organic coating-metal interface? 3. What is the thickness of the water layer at the interface, and what deter- mines this thickness? Two additional questions could be added to this list: 4. What makes adhesion loss under wet circumstances irreversible? Is there a relationship between the coating property, wet adhesion, the failure mechanism, and water disbondment? 5. Why does the reduction of adhesion on exposure to water not lead to complete delamination? What causes residual adhesion in wet circum- stances? As a possible answer to the last question above, Funke has suggested that dry adhesion is due to a mixture of bond types. Polar bonding, which is somewhat sensitive to water molecules, could account for reduced adhesion in wet circum- stances, whereas chemical bonds or mechanical locking may account for residual adhesion [16]. Further research on wet adhesion could answer some of the afore- mentioned questions and increase understanding of this complex mechanism. 1.2.3.3 Important Aspects of Adhesion Two aspects of adhesion are important: the initial strength of the coating-substrate bond and what happens to this bond as the coating ages. 7278_C001.fm Page 7 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC 8 Corrosion Control Through Organic Coatings A great deal of work has been done to develop better methods for measuring the initial strength of the coating-substrate bond. Unfortunately, the emphasis on measuring initial adhesion may miss the point completely. It is certainly true that good adhesion between the metal and the coating is necessary for preventing cor- rosion under the coating. However, it is possible to pay too much attention to measuring the difference between very good initial adhesion and excellent initial adhesion, completely missing the question of whether or not that adhesion is main- tained. In other words, as long as the coating has good initial adhesion, then it does not matter whether that adhesion is simply very good or great. What matters is what happens to the adhesion over time. This aspect is much more crucial to coating success or failure than is the exact value of the initial adhesion. Adhesion tests on aged coatings are useful not only to ascertain if the coatings still adhere to the metal but also to yield information about the mechanisms of coating failure. This area deserves greater attention, because studying changes in the failure loci in adhesion tests before and after weathering can yield a great deal of information about coating deterioration. 1.2.4 P ASSIVATING WITH P IGMENTS Anticorrosion pigments in a coating dissolve in the presence of water. Their dissociated ions migrate to the coating-metal interface and passivate it by supporting the formation of thin layers of insoluble corrosion products, which inhibit further corrosion [44-46]. For more information about anticorrosion pigments, see Chapter 2. 1.2.5 A LTERNATIVE A NODES (C ATHODIC P ROTECTION ) Some very effective anticorrosion coatings allow the conditions necessary for cor- rosion to occur — for example, water, oxygen, and ions may be present; the coating does not offer much electrical resistance; or soluble pigments have not passivated the metal surface. These coatings do not protect by suppressing the corrosion process; rather, they provide another metal that will corrode instead of the substrate. This mechanism is referred to as cathodic protection. In protective coatings, the most important example of cathodic protection is zinc-rich paints, whose zinc pigment acts as a sacrificial anode, corroding preferentially to the steel substrate. For more information on zinc-rich coatings, see Chapter 2. REFERENCES 1. Koehler, E.L., Corrosion under organic coatings, Proc., U.R Evans International Conference on Localized Corrosion, NACE, Houston, 1971, 117. 2. Guruviah, S., JOCCA, 53, 669, 1970. 3. Mayne, J.E.O., JOCCA, 32, 481, 1949. 4. Thomas, A.M and Gent, W.L., Proc. Phys. Soc., 57, 324, 1945. 5. Anderson, A.P. and Wright, K.A., Industr. Engng. Chem., 33, 991, 1941. 6. Edwards, J.D. and Wray, R.I., Industr. Engng. Chem., 28, 549, 1936 7. Maitland, C.C. and Mayne, J.E.O., Off. Dig., 34, 972, 1962. 8. McSweeney, E.E., Off. Dig., 37, 626, 1965. 7278_C001.fm Page 8 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC Introduction 9 9. Thomas. N.L., Prog. Org. Coat., 19, 101, 1991. 10. Thomas, N.L., Proc. Symp. Advances in Corrosion Protection by Organic Coatings, Electrochem. Soc., 1989, 451. 11. Haagen, H. and Funke, W., JOCCA, 58, 359. 1975. 12. Wheat, N., Prot. Coat. Eur., 3, 24, 1998. 13. Khullar, M.L. and Ulfvarson, U., Proc., IXth FATIPEC Congress, Fédération d’Asso- ciations de Techniciens des Industries des Peintures, Vernis, Emaux et Encres d’Imprimerie de l’Europe Continentale (FATIPEC), Paris, 1968, 165. 14. Bacon, C. et al., Ind. Eng. Chem., 161, 40, 1948. 15. Cherry, B.W., Australag. Corr. and Eng., 10, 18, 1974. 16. Funke, W., in Surface Coatings – 2, Wilson, A.D., Nicholson, J.W. and Prosser, H.J., Eds., Elsevier Applied Science, London, 1988, 107. 17 Kaesche, H., Werkstoffe u. Korrosion, 15, 379, 1964. 18. Knotkowa-Cermakova, A. and Vlekova, J., Werkstoffe u. Korrosion, 21, 16, 1970. 19. Schikorr, G., Werkstoffe u. Korrosion, 15, 457, 1964. 20. Dunkan, J.R., Werkstoffe u. Korrosion, 25, 420, 1974. 21. Barton, K. and Beranek, E., Werkstoffe u. Korrosion, 10, 377, 1959. 22. Kittleberger, W.W. and Elm, A.C. Ind. Eng. Chem., 44, 326, 1952. 23. Bacon, C.R., Smith, J.J. and Rugg, F.M., Ind. Eng. Chem., 40, 161, 1948. 24. Cherry, B.W. and Mayne, J.E.O., Proc., First International Congress on Metallic Corrosion, Butterworths, London, 1961. 25. Mayne, J.E.O., Trans. Inst. Met. Finish., 41, 121, 1964. 26. Cherry, B.W. and Mayne, J.E.O. Off. Dig., 37, 13, 1965. 27. Mayne, J.E.O., JOCCA, 40, 183, 1957. 28. Ulfvarson, U. and Khullar, M., JOCCA, 54, 604, 1971. 29. Walter, G.W., Corros. Science, 26, 27, 1986. 30. Kumins, C.A. and London, A., J. Polymer Science, 46, 395, 1960. 31. Greenfield, D. and Scantlebury, D., J. Corros. Science and Eng., 3, Paper 5, 2000. 32. Patrick, R.L. and Millar, R.L. in Handbook of Adhesives, Skeist, I., Ed. Reinhold Publishing Corp., New York, 1962, 602. 33. Kittleberger, W.W. and Elm, A.C., Ind. Eng. Chem., 38, 695, 1946. 34. Evans, U.R. Corrosion and Oxidation of Metals, St. Martin’s Press, New York. 1960. 35. Gowers, K.R. and Scantlebury, J.D. JOCCA, 4, 114, 1988. 36. Troyk, P.R., Watson, M.J. and Poyezdala, J.J., in ACS Symposium Series 322: Poly- meric Materials for Corrosion Control, Dickie, R.A. and Floyd, F.L, Eds., American Chemical Society, Washington DC, 1986, 299. 37. Bullett, T.R., JOCCA, 46, 441, 1963. 38. Walker, P., Off. Dig., 37, 1561, 1965. 39. Walker, P., Paint Technol., 31, 22, 1967. 40. Walker, P., Paint Technol., 31, 15, 1967. 41. Funke, W., J. Coat. Technol., 55, 31, 1983. 42. Funke, W., in ACS Symposium Series 322: Polymeric Materials for Corrosion Control, Dickie, R.A. and Floyd, F.L, Eds., American Chemical Society, Washington DC, 1986, 222. 43. Leidheiser, H., in ACS Symposium Series 322: Polymeric Materials for Corrosion Control, Dickie, R.A. and Floyd, F.L, Eds., American Chemical Society, Washington DC, 1986, 124. 44. Mayne, J.E.O. and Ramshaw, E.H. J. Appl. Chem., 13, 553, 1969. 45. Leidheiser, H., J. Coat. Technol., 53, 29, 1981. 46. Mayne, J.E.O., in Pigment Handbook, Vol. III, Patton, T.C., Ed. Wiley Interscience Publ., 1973, 457. 7278_C001.fm Page 9 Friday, February 3, 2006 12:34 PM © 2006 by Taylor & Francis Group, LLC . good corrosion protection. For all of the protection mechanisms described in the previous sections, good adhesion of the coating 727 8_C001.fm Page 6 Friday, February 3, 20 06 12: 34 PM © 20 06. Chem., 28 , 549, 1936 7. Maitland, C.C. and Mayne, J.E.O., Off. Dig., 34, 9 72, 19 62. 8. McSweeney, E.E., Off. Dig., 37, 626 , 1965. 727 8_C001.fm Page 8 Friday, February 3, 20 06 12: 34. 1986, 22 2. 43. Leidheiser, H., in ACS Symposium Series 322 : Polymeric Materials for Corrosion Control, Dickie, R.A. and Floyd, F.L, Eds., American Chemical Society, Washington DC, 1986, 124 . 44.